I don't know if this is a theoretical draw or win but that doesn't affect the moves I'd play. I'd try to get the black king to e3 and play a back line check, after which I could maybe get the K to e2. Then perhaps I could get the R to f1 and then f2 and the K to f1. The idea would be to zugzwang the N and reach a position where the enemy's pawn can be won.
No idea if this would work, but I'd try it.
Looking at the tablebase might help - go to chessok.com and input the position. Shows mate in 23, and provides the full line where white puts up the longest defense. If you have questions about what happens if white varies, that is easy to check. This is of much more help than an engine (for positions which are within the tablebases) -- my stockfish could not solve this position before I lost patience.
I'd try to get the black king to e3 and play a back line check, after which I could maybe get the K to e2. Then perhaps I could get the R to f1 and then f2 and the K to f1.
No idea if this would work, but I'd try it.
Thank you very much!
Your analysis matches Komodo 9's finding. Stockfish 8 couldn't solve it in a timely matter, don't know if it could do it with more time.
ArtNJ:
Thank you as well for the tip about using that free services. I knew about the tablebases but wasn't aware there was a free service with them ready to use. I see you try stockfish too.... well, it made feel not that bad that a "famous" engine couldn't solve it either.
Thank you, guys !
